Effect of a 9-Month Internship Intervention for Military Dependents With ASD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of a 9-Month Internship Intervention for Military Dependents With ASD (NCT04001790), a randomized na study. The primary outcome is Change in Competitive Integrated Employment Rates.

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group trial with no masking. One group received Project Search plus ASD Supports; the comparison group received Business as Usual. Random assignment is relevant because, when implemented successfully, it can support a fair comparison of groups by reducing systematic differences at the point of allocation. The record does not report how the random sequence was generated, how assignment was concealed, or whether allocation was stratified.

The listed enrollment was 48. The posted primary outcome lists 20.0 participants in the Project Search plus ASD Supports group and 18.0 in Business as Usual. These are the group denominators associated with the reported outcome, and the record does not explain why they differ from the enrollment figure. The brief summary separately describes recruitment activities and cohort participation, but does not provide a reconciliation between those details and the posted outcome denominators.

The intervention was designed around an intensive 9-month employment program, with youth placed in a community business for internship rotations. Business as Usual participants remained in high school programs shaped by individualized education plans. The registry says that the project modified the Project SEARCH Plus ASD Model for military dependents with ASD, measured implementation fidelity, and intended to assess social communication, behavioral adjustment, and employment outcomes. Only one outcome measure is posted, and no secondary outcomes are listed.

The trial status is completed. The registry lists a start date of 2017-05-11 and a primary completion date of 2021-05-19. Details about follow-up procedures, protocol deviations, and the conduct of the trial are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Employed at 18 months

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Change in Competitive Integrated Employment Rates, measured over baseline to 18 months.

The primary endpoint is titled “Change in Competitive Integrated Employment Rates.” Its description concerns participants engaged in full- or part-time work at minimum wage or higher, with wages and benefits accorded to non-disabled workers in the same position and integration with co-workers and customers without disabilities. The timeframe is baseline to 18 months, and the primary population is described as “Employed at 18 months.”

This endpoint focuses on whether participants enter employment that meets the registry’s criteria, rather than simply whether they have any work arrangement. Employment is a direct outcome for an intervention intended to provide job training and internships. In general, a prespecified endpoint helps define what a trial is intended to compare and can keep interpretation focused on an outcome selected before results are known. The registry does not report why this endpoint was chosen over other possible outcomes or how each employment criterion was verified.

The posted result is expressed as a count of participants: 12.0 in the intervention group and 0.0 in Business as Usual. Although the title refers to change and the timeframe begins at baseline, baseline employment counts and individual transitions into or out of employment are not reported in the registry record. The posted information therefore describes the reported group counts at the stated outcome time; it does not show the participant-level paths that produced those counts.

7. Results Posted in the Registry

• Project Search + ASD Supports: 60.0% (12/20) versus 0.0% (0/18) on control; risk difference 60.0 percentage points (95% CI 38.5 to 81.5); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a risk difference describe?

A risk difference is an absolute contrast between the proportion with an outcome in one group and the proportion with the outcome in another. It keeps the comparison in the same units as the underlying risks: a difference in the probability of the event. General statistical interpretation: an absolute measure can be useful when readers need to understand how far apart the groups are on the outcome scale, while a relative measure answers a different question about proportional change.

For this endpoint, the posted group risks are 0.6 for Project Search plus ASD Supports and 0.0 for Business as Usual. The registry reports their risk difference as 0.6. The record labels the primary measure as a count of participants, and also reports a comparison on the risk-difference scale. These are related but not identical presentations: counts retain the number of participants with the outcome, while risks account for the group denominators. Interpretation should keep the endpoint definition and each group’s reported denominator in view.

What does the p-value tell the reader?

The registry reports an unadjusted p-value of p < 0.001. In general, a p-value describes how unusual results at least as incompatible with a null hypothesis would be if that null model and the statistical assumptions were correct. It does not give the probability that the result occurred by chance, prove that a treatment effect exists, or convey the size of an effect. The effect estimate and its uncertainty are needed for those parts of interpretation.

The registry record does not report the test used to obtain this p-value, whether it was one-sided or two-sided, or what adjustments, if any, were made for design features. The displayed label says unadjusted, so it should not be described as adjusted for other variables or for multiple comparisons. The record does not list a statistical analysis plan that would clarify the intended test or analysis hierarchy.

Why is the confidence interval important?

A confidence interval gives a range of estimates compatible with the observed data under the procedure used to construct it. In general, a narrower interval indicates greater precision for the estimated effect, while a wider interval indicates greater uncertainty. It is not a guarantee that the true effect lies in the interval, and its interpretation depends on the method and assumptions used.

The registry reports an interval for the risk difference, with both limits above zero. This conveys uncertainty around the point estimate as well as the direction of the reported contrast. The method used to construct the interval is not reported in the registry record. The interval should be considered with the group counts, outcome definition, and limitations in follow-up and analysis reporting, rather than treated as a complete description of uncertainty.

What can be learned from the group counts?

The group counts make the basis of the comparison more concrete: 12.0 participants were recorded with the outcome in the intervention group, and 0.0 in Business as Usual. The denominators attached to those results are 20.0 and 18.0. General statistical principle: when outcomes are binary, the number with the event and the number evaluated are both important. A count alone cannot convey the probability of the outcome without its denominator, and a risk alone can obscure how many observations support it.

The outcome is described as a count, not as a time-to-event measure. The registry does not report event dates, time to employment, or an analysis of time until employment. It also does not provide participant-level data or a breakdown by employment status at baseline. Consequently, the posted result cannot establish when employment began or whether participants remained employed throughout follow-up.

How should the planning illustration be read?

The registry includes assumptions for a planning illustration based on the observed effect. The stated assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. These assumptions describe an illustration, not necessarily the design process used for this trial. The sample-size justification and the trial’s original planning assumptions are not reported in the registry record.

Any sample sizes in such a planning exercise are prospective illustrations, not observed power and not evidence that the completed trial had a particular probability of detecting an effect. The illustration depends on the effect chosen, the assumed significance level, allocation, and approximation. It should not be taken as a substitute for uncertainty intervals around the actual result. It also does not establish that the trial was adequately powered for other outcomes, such as social communication or behavioral adjustment.

The record does not list a discrepancy between the posted values and a sponsor analysis. The statistical method used for the sponsor’s primary analysis is not reported in the registry record, and the record does not describe whether the presented comparison was prespecified. Without those details, interpretation should remain anchored to the posted estimate, interval, p-value, and outcome counts rather than assuming unreported analytic choices.

10. Confidence Intervals and Precision

The registry reports a risk difference of 0.6 and a confidence interval whose limits are both above zero. The interval is the principal posted summary of precision for the estimated contrast. It indicates that the point estimate should not be read as exact: the underlying effect could differ from the estimate, even if the observed direction is consistent with the reported interval.

In general, confidence intervals communicate the range of values that remain compatible with the data under a specified statistical method. They give more information than a p-value alone because they show the scale of uncertainty around the effect estimate. A p-value below a conventional threshold can indicate evidence against a null model, but it does not show whether the estimate is precise enough for a particular decision. The registry does not report the interval construction method, so the assumptions behind its coverage cannot be assessed from the record.

The result comes from a relatively small set of posted outcome denominators, 20.0 and 18.0. With binary outcomes, a limited number of participants can make estimates sensitive to the outcomes of individual participants. This is a general consideration, not a claim that any specific participant would have changed the conclusion. The registry does not report sensitivity analyses to show how robust the estimate is to alternative assumptions or outcome classifications.

Precision is also distinct from generalizability. Even a precise estimate would apply most directly to settings and participants resembling those studied, while the registry record provides limited detail about participant characteristics, sites, and implementation. Conversely, uncertainty about applicability is not summarized by the confidence interval. The interval addresses statistical uncertainty in the estimated contrast under its method, not every source of uncertainty about the program or its delivery.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one posted outcome measure and no posted secondary outcomes. It lists no analyses. This does not establish that no additional analyses were conducted; it means that analysis details beyond the posted result are not reported in the registry record. The project description mentions aims involving social communication, behavioral adjustment, employment, and implementation fidelity, but the registry does not post results for those other aims as outcome measures.

Multiplicity arises when a study tests more than one outcome, comparison, time point, or subgroup. In general, repeated testing can increase the chance of finding at least one apparently unusual result even when the null model is true. Methods for addressing that issue depend on the analysis plan and the study’s goals. The record does not report a multiplicity strategy, a hierarchy of outcomes, or whether the posted primary result was selected from a larger set of analyses. The p-value is labeled unadjusted and should be understood accordingly.

Missing-data procedures are not reported in the registry record. The posted outcome denominators are 20.0 for the intervention group and 18.0 for Business as Usual, while the listed enrollment is 48. The record does not explain the relationship among these figures, identify reasons for absent outcomes, or say whether all randomized participants were included in the analysis. It also does not report whether missing outcomes were imputed, treated as no employment, or handled in another way.

The analysis population is not described. The record does not say whether the comparison followed an intention-to-treat approach, included only participants who completed the intervention, or used another defined population. It also does not report adherence or crossovers. These omissions matter because analysis choices can change the meaning of a randomized comparison, particularly when outcome availability or participation differs between groups. They prevent a complete assessment of bias related to missingness and post-randomization exclusions.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry reports serious adverse events, other adverse events, and deaths separately for the two groups. For Project Search plus ASD Supports, the serious-event group lists 0 events among 22 at risk, the other-event group lists 0 among 22 at risk, and deaths are listed as 0. For Business as Usual, the serious-event group lists 0 among 26 at risk, the other-event group lists 0 among 26 at risk, and deaths are listed as 0. No serious-event terms are listed.

These are the safety figures posted in the registry. The numbers described as at risk are not the same as the denominators attached to the employment outcome, and the registry does not explain how the safety populations were defined or how they relate to the outcome populations. The figures should therefore be read within their own reported categories rather than combined with the employment results.

Zero reported events means that no events are recorded in these registry fields; it does not establish that no adverse experiences occurred or that the intervention is free of risk. The record does not provide event definitions, collection procedures, severity details beyond the serious/other categories, or information about completeness of reporting. It also does not describe how safety events were reviewed or attributed. These limits constrain what can be concluded from the posted safety counts.

13. What the Result Does, and Does Not, Mean

The result documents a difference in the posted employment outcome counts between the randomized groups: 12.0 participants in Project Search plus ASD Supports were recorded with the outcome, compared with 0.0 in Business as Usual. The registry reports a risk difference of 0.6 and an unadjusted p-value of p < 0.001. The interval limits are both above zero, while still conveying uncertainty around the point estimate. Together, these figures describe a strong observed separation in this particular comparison.

The result does not establish that every participant in the intervention group benefited, that the effect would be the same in other populations or settings, or that the program alone caused each employment outcome. Randomization is an important design feature, but the registry does not report allocation procedures, missing-data handling, analysis populations, or outcome assessment methods. The absence of masking may also matter if knowledge of assignment could influence implementation or classification of employment.

The result does not show the timing or durability of employment. Although the primary timeframe is baseline to 18 months and the population is described as employed at 18 months, the record does not give baseline counts, individual employment transitions, or a time-to-employment analysis. Nor does the posted outcome describe job quality beyond the stated employment criteria or report outcomes such as social communication and behavioral adjustment.

The record also does not establish that the same effect would be achieved under different delivery conditions. The project aimed to adapt the model for military dependents and measure fidelity, but fidelity results and details about implementation are not posted as outcome measures. The employment comparison should therefore be interpreted as the reported result for this trial and endpoint, with the limits of the registry information in mind—not as a complete evaluation of every project aim.

14. Why This Trial Matters Statistically

This trial is statistically notable because it compares a structured employment intervention with a defined usual-program comparator using random allocation and a binary employment outcome. A risk difference gives an absolute comparison that can be read alongside the group counts. The posted result reports 12.0 participants with the endpoint in Project Search plus ASD Supports and 0.0 in Business as Usual, together with a risk difference of 0.6. That combination makes the scale and direction of the recorded contrast more interpretable than a p-value alone.

The trial also illustrates why an endpoint’s label and operational definition deserve close attention. The title refers to change in employment rates, while the posted population is “Employed at 18 months” and the outcome is recorded as a count. Without baseline counts or transition data, the posted information does not show how change over time was calculated. Clear alignment between endpoint wording, time frame, and reported analysis is important for understanding what a result actually estimates.

Several statistical reporting details are essential for judging the result but are not reported in the registry record: the analysis population, missing-data methods, outcome assessment procedures, and the statistical test behind the p-value and interval. The listed outcome denominators also differ from the enrollment figure, without an explanation. These gaps show why a striking point estimate is not the whole statistical story. Precision, data completeness, and analysis choices all shape confidence in an estimated effect.

The trial’s relevance is also tied to its specific intervention and comparator. The program involved workplace internships, whereas Business as Usual kept youth in high school programs based on individualized education plans. Statistical interpretation is strongest when it stays connected to those real group definitions instead of treating the result as a generic comparison of employment programs.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record leaves important details about the trial population unreported. It identifies military dependents with ASD as the focus and describes youth with developmental disabilities in their last year of high school, but does not provide further eligibility criteria, participant characteristics, or the settings represented in the outcome. Those omissions limit assessment of how well the results may apply to other groups or service contexts.

The primary endpoint is described as a change from baseline to 18 months, but the posted outcome is a count of participants employed at 18 months. Baseline employment values and participant-level transitions are not reported in the registry record. The record also does not explain how employment criteria were verified, who assessed the outcome, or whether assessors were aware of group assignment. These details matter when interpreting a classification-based endpoint.

The registry lists enrollment of 48 and outcome denominators of 20.0 and 18.0, but does not reconcile them. Missingness, exclusions, adherence, crossover, and analysis population are not reported. The statistical test and confidence-interval method are also unreported. As a result, the posted estimate cannot be fully evaluated for sensitivity to attrition, analysis choices, or assumptions.

The trial was conducted without masking. That may be relevant to intervention delivery and outcome assessment, although the record does not document that bias occurred. The project description includes goals beyond employment, such as social communication, behavioral adjustment, fidelity, and relocation or deployment planning, but corresponding outcome results are not posted. Safety reporting lists zero events in the posted categories, but gives limited information about collection and ascertainment. These are limitations of what can be concluded from the registry record, not proof that problems occurred.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Effect of a 9-Month Internship Intervention for Military Dependents With ASD — NCT04001790.
• Registry record last updated 2024-12-09; status last verified 2024-10.
• Results first posted 2024-12-09.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT04001790 is a completed, randomized, parallel-group trial with no masking, comparing Project Search plus ASD Supports with Business as Usual for military dependents with ASD. The intervention was an intensive 9-month job training program involving internships; the comparator kept youth in high school programs according to individualized education plans.

The primary outcome concerns competitive integrated employment at 18 months. The posted counts are 12.0 participants in the intervention group and 0.0 in Business as Usual, with outcome denominators of 20.0 and 18.0. The registry reports a risk difference of 0.6, an unadjusted p-value of p < 0.001, and a confidence interval whose limits are above zero.

One outcome measure is posted, and no analyses are listed. Details about the statistical test, analysis population, missing-data handling, and employment assessment are not reported in the registry record. Safety fields list zero serious adverse events, other adverse events, and deaths in both groups, with separate at-risk counts.